Releases: intel/openvino-plugins-ai-audacity
v3.7.0-R4.0
Compatibility 🔧
Changes since previous release:
- Audacity compatibility updated from 3.6.4 to 3.7.0
- Fixed issue that was causing problems when Windows username has special characters (#313, #293) -- Thank you @Courteous-Cheetah for helping to validate!
⚠️ If you've previously installed any of our 3.6.X or 3.5.X plugin releases, the recommendation for upgrade is to not uninstall the previous version. Instead, just download this version and select Install no models from the drop down list. This way, you won't need to re-download all of the models again.
🐧 Although we are only providing an installer for Windows right now, it can be compiled for Linux using this build procedure.
Supported Hardware
As this release is built on top of OpenVINO™ 2024.3.0, it will generally support hardware defined in this list: https://docs.openvino.ai/2024/about-openvino/release-notes-openvino/system-requirements.html
Need help? 🙋
If you encounter any issues during install procedure, reach out to us for help by filing an issue here. We are happy to help!
Contributors
Assets 3
v3.6.4-R3.4
Compatibility 🔧
Changes since previous release:
- Audacity compatibility updated from 3.6.3 to 3.6.4
⚠️ If you've previously installed any of our 3.6.X or 3.5.X plugin releases, the recommendation for upgrade is to not uninstall the previous version. Instead, just download this version and select Install no models from the drop down list. This way, you won't need to re-download all of the models again.
🐧 Although we are only providing an installer for Windows right now, it can be compiled for Linux using this build procedure.
Supported Hardware
As this release is built on top of OpenVINO™ 2024.3.0, it will generally support hardware defined in this list: https://docs.openvino.ai/2024/about-openvino/release-notes-openvino/system-requirements.html
Need help? 🙋
If you encounter any issues during install procedure, reach out to us for help by filing an issue here. We are happy to help!
v3.6.3-R3.3
Compatibility 🔧
Changes since previous release:
- Audacity compatibility updated from 3.6.2 to 3.6.3
⚠️ If you've previously installed any of our 3.6.X or 3.5.X plugin releases, the recommendation for upgrade is to not uninstall the previous version. Instead, just download this version and select Install no models from the drop down list. This way, you won't need to re-download all of the models again.
🐧 Although we are only providing an installer for Windows right now, it can be compiled for Linux using this build procedure.
Supported Hardware
As this release is built on top of OpenVINO™ 2024.3.0, it will generally support hardware defined in this list: https://docs.openvino.ai/2024/about-openvino/release-notes-openvino/system-requirements.html
Need help? 🙋
If you encounter any issues during install procedure, reach out to us for help by filing an issue here. We are happy to help!
v3.6.2-R3.2
Compatibility 🔧
Changes since previous release:
- Audacity compatibility updated from 3.6.1 to 3.6.2
- Updated libtorch to 2.4.1
- Updated OpenVINO & OpenVINO Tokenizers to 2024.3.0
⚠️ If you've previously installed any of our 3.6.X or 3.5.X plugin releases, the recommendation for upgrade is to not uninstall the previous version. Instead, just download this version and select Install no models from the drop down list. This way, you won't need to re-download all of the models again.
🐧 Although we are only providing an installer for Windows right now, it can be compiled for Linux using this build procedure.
Supported Hardware
As this release is built on top of OpenVINO™ 2024.3.0, it will generally support hardware defined in this list: https://docs.openvino.ai/2024/about-openvino/release-notes-openvino/system-requirements.html
Need help? 🙋
If you encounter any issues during install procedure, reach out to us for help by filing an issue here. We are happy to help!
v3.6.1-R3.1
Compatibility 🔧
Changes since previous release:
- Audacity compatibility updated from 3.6.0 to 3.6.1
⚠️ If you've previously installed our 3.6.0, or any of our previous 3.5.X plugin releases, the recommendation for upgrade is to not uninstall the previous version. Instead, just download this version and select Install no models from the drop down list. This way, you won't need to re-download all of the models again.
🐧 Although we are only providing an installer for Windows right now, it can be compiled for Linux using this build procedure.
Supported Hardware
As this release is built on top of OpenVINO™ 2024.2.0, it will generally support hardware defined in this list: https://docs.openvino.ai/2024/about-openvino/release-notes-openvino/system-requirements.html
Need help? 🙋
If you encounter any issues during install procedure, reach out to us for help by filing an issue here. We are happy to help!
v3.6.0-R3.0
Compatibility 🔧
Changes since previous release:
- OpenVINO has been updated to latest version, 2024.2.
- Changes to support updated WaveTrack APIs since 3.5.X.
⚠️ If you've previously installed any of our 3.5.X plugin releases, the recommendation for upgrade is to not uninstall the previous version. Instead, just download this version and select Install no models from the drop down list. This way, you won't need to re-download all of the models again.
🐧 Although we are only providing an installer for Windows right now, it can be compiled for Linux using this build procedure.
Supported Hardware
As this release is built on top of OpenVINO™ 2024.2.0, it will generally support hardware defined in this list: https://docs.openvino.ai/2024/about-openvino/release-notes-openvino/system-requirements.html
Need help? 🙋
If you encounter any issues during install procedure, reach out to us for help by filing an issue here. We are happy to help!
v3.5.1-R2.2
v3.5.1-R2.2 is a patch release that includes the following improvements:
Fixes:
- Model Caching issues on non-English locales (such as German, French, etc.), which caused features (such as Music Separation) to only work the first time that they were used. #1
- Unable to Load "mod-openvino": The specified module cannot be found. Due to lack of OpenCL.dll #182, #141.
- Transcription fails if run from Macro. #160
Whisper transcription improvements:
- Languages are now listed in alphabetical order. #186
- Audacity no longer goes to 'Not Responding' state in-between progress updates when larger models are used. #185
- Underlying whisper.cpp has been updated to v1.6.0, which should resolve some issues such as this one: #187
⚠️ If you've previously installed 3.5.0-R2, 3.5.1-R2, or 3.5.1-R2.1 release, the recommendation for upgrade is to not uninstall the previous version. Instead, just download this version and select Install no models from the drop down list. This way, you won't need to re-download all of the models again.
Compatibility 🔧
🐧 Although we are only providing an installer for Windows right now, it can be compiled for Linux using this build procedure.
Supported Hardware
As this release is built on top of OpenVINO™ 2024.0, it will generally support hardware defined in this list: https://docs.openvino.ai/2024/about-openvino/release-notes-openvino/system-requirements.html
Need help? 🙋
If you encounter any issues during install procedure, reach out to us for help by filing an issue here. We are happy to help!
v3.5.1-R2.1
v3.5.1-R2.1 is a patch release that fixes the following issues:
- Whisper Transcription fails with
invalid map<K, T> key. Reported in many places, but tracked here: #143 - Installer throws
Failed to set directory permissions. Issues: #157, #146, #140
⚠️ If you've previously installed 3.5.0-R2, or 3.5.1-R2 release, the recommendation for upgrade is to not uninstall the previous version. Instead, just download this version and select Install no models from the drop down list. This way, you won't need to re-download all of the models again.
Compatibility 🔧
🐧 Although we are only providing an installer for Windows right now, it can be compiled for Linux using this build procedure.
Supported Hardware
As this release is built on top of OpenVINO™ 2024.0, it will generally support hardware defined in this list: https://docs.openvino.ai/2024/about-openvino/release-notes-openvino/system-requirements.html
Need help? 🙋
If you encounter any issues during install procedure, reach out to us for help by filing an issue here. We are happy to help!
v3.5.1-R2
This is a hotfix release that adds compatibility for Audacity 3.5.1. There are no other changes vs. our previous 3.5.0-R1 release.
-
At the Select Models step, choose Install no models from the drop-down list. This way, you won't need to re-download all of the models again:
-
A Components Exist warning window will pop up -- click Yes.
Compatibility 🔧
🐧 Although we are only providing an installer for Windows right now, it can be compiled for Linux using this build procedure.
Supported Hardware
As this release is built on top of OpenVINO™ 2024.0, it will generally support hardware defined in this list: https://docs.openvino.ai/2024/about-openvino/release-notes-openvino/system-requirements.html
Need help? 🙋
If you encounter any issues during install procedure, reach out to us for help by filing an issue here. We are happy to help!
v3.5.0-R2
🎉 We are thrilled to share this new release of OpenVINO™ AI Plugins for Audacity*! It includes a bunch of improvements over the initial release, and is compatible with Audacity 3.5.0! 🤘
💾 Download the installer here.
Compatibility 🔧
🐧 Although we are only providing an installer for Windows right now, it can be compiled for Linux using this build procedure.
Major changes
- We now have an installer for windows! During installation, you can select which models to install. The installer will take care of downloading the required model files and installing them into the right system location.
- Music Generation now uses MusicGen LLM (from Meta), and now supports Music Continuation. Take a look at the documentation page which include a couple of short videos that demonstrate Music Generation & Continuation features.
- Noise Suppression now has support for DeepFilter2, and DeepFilter3 models. These models run much faster and produce higher quality results than the denseunet model that was included in the previous release.
- Whisper Transcription supports small, medium, and large models. The model selection drop-down list is automatically populated with the models that the user chose to install.
Other changes
Additions
- All effects have a Device Info button added to which gives some more verbose information about the devices listed in the drop-down list.
- Whisper Transcription has some 'advanced options' added such as Initial Prompt, Max Segment Length, and some others. See documentation for more info.
- Music Separation has Shifts parameter exposed as an Advanced Option. Note that in previous release, this was hard-coded to 2. See documentation for more info.
- Noise Suppression has some Advanced Options available, based on the current model selection. See documentation for more info.
- The version of OpenVINO™ that is bundled with this module has been upgraded to 2024.0 release
Removals
- Riffusion-based effects (original Music Generation, and Music Style Remix) have been removed.
Fixes
- Module load issues on x86 systems that do not support AVX / AVX2 should be resolved. #8
- Non-selected audio is no longer affected by running Noise Suppression. #69
- Usernames with non-ascii characters should no longer trigger OpenVINO permission errors. #114 #45
Supported Hardware
As this release is built on top of OpenVINO™ 2024.0, it will generally support hardware defined in this list: https://docs.openvino.ai/2024/about-openvino/release-notes-openvino/system-requirements.html
Known Issues & Limitations
- The installer may default to an Audacity installation which is not supported by these plugins.
- When using Intel® Core™ Ultra systems, there are some known limitations when using NPU:
- Note: All testing was performed using
31.0.100.2016version of NPU driver. - Whisper transcription works for base, small, and medium model variants, but runtime errors are encountered for Large models.
- Music Separation (demucs v4) model is not yet supported by NPU.
- The first time using a feature for NPU, the Model Load time may be significantly longer than for other devices.
- Note: All testing was performed using
🙏 A huge thanks to everyone who tried out, reported bugs, and gave feedback on the first release!!
Need help? 🙋
If you encounter any issues during install procedure, reach out to us for help by filing an issue here. We are happy to help!